Background

Lithium-ion batteries underpin much of the modern world, from their use in mobile phones to electric vehicles and large-scale energy storage. With the increase in use, the safety of batteries becomes more of a priority than ever. Batteries age very differently depending on their chemistry, shape, size, arrangement, and the typical environment they operate in. It is vital, therefore, to understand the effect of time and multiple charge-discharge cycles on the battery's physical, chemical and performance properties. Ageing batteries behave differently from new ones and become more susceptible to faults.

Challenge

A major challenge for gathering large volumes of high-quality data is that battery ageing experiments are resource intensive and can take anywhere from six months to two years. To accurately predict the ageing of the battery and the effect on safety, the testing conditions must mimic real-life working conditions as closely as possible.

Manufacturers increasingly need to understand how and when batteries age and how the performance of a battery will be affected over time. The key challenge is trying to condense a normal lifetime's worth of charge-discharge cycles into a few weeks rather than a year, without producing misleading results. For the UK to establish a thriving and robust battery manufacturing industry, it was necessary that facilities were established to investigate the long-term ageing of different cell types without interruption.

Solution

With funding from Faraday, Warwick Manufacturing Group (WMG) established a laboratory at the Wellesbourne Campus. This unique UK facility can provide a dedicated analysis of the problem of precision-ageing of lithium-ion battery cells. The High Value Manufacturing Catapult established a dedicated team of researchers, engineers and technicians for the facility and set up bespoke test methodologies with immersed thermal management rigs. Testing produced substantial amounts of data. This data and that from studies at WMG's Energy Innovation Centre had to all be collated, stored, accessed, and analysed on a scale not previously attempted at WMG.

The work resulted in the creation of an important repository of reference datasets on lithium-ion battery ageing and degradation, which can benefit the UK battery industry. Researchers at WMG have produced fully validated and parameterised high-accuracy ageing and degradation models. The reference dataset is intended to support and accelerate the development of new machine learning and artificial intelligence (AI) battery ageing algorithms. It has also led to the generation of new knowledge and a better understanding of electrochemical ageing mechanisms through forensic autopsy and physical validation of ageing mechanisms.

The data and modelling have benefitted UK battery manufacturers, automotive companies producing electric vehicles, and other sectors seeking to move to electrification. A Collaborative Research & Development project with Innovate UK named project COBRA (Cloud/Onboard Battery Remaining useful life Algorithm), was a £169K project led by Eatron in collaboration with WMG. It developed state-of-the-art battery Remaining Use Life (RUL) prediction algorithms as part of its Battery Management System technology solution.

"The Warwick Wellesbourne testing facility has enabled Eatron to access valuable data on battery cell performance as well as expert advice from leading academics. As a growing SME, this has saved Eatron the significant cost of building and maintaining our own testing facilities, allowing us to focus on core competencies in software and algorithms for battery management. Our joint activities with WMG have helped Eatron to optimise our BMSTAR battery management system and stay at the forefront of battery technology."

Eatron Technologies

Academic papers are being generated for the new methodologies developed at the facility using the immersed thermal management rig, "Novel Techniques for Economical and Efficient Large-Scale Testing of Lithium-ion Batteries using Improved Isothermal Control". Also, "Improving Model Parameterisation and Accuracy through Isothermal Testing of Lithium-ion Batteries for Automotive/Aerospace Applications." These papers provide open-access knowledge resources for rig design and processes.

Impact

Ageing data is now freely available to any research or academic organisation. The University of Edinburgh published a journal entitled "Lithium-ion battery data and where to find it" which highlights the WMG battery data set. WMG have created a Data Sharing Agreement with several UK Universities to provide access to the data. This data sharing promotes a collaborative approach to enhancing a detailed understanding of battery ageing.

With WMG providing this distinctive facility in the UK and being able to speed up development times for new batteries, there has been a huge impact on the UK automotive sector. The High Value Manufacturing Catapult also invested in compiling data from the ageing studies into a database accessible to researchers. A Graphical User Interface (GUI) was created to enable WMG's researchers to search for datasets by predefined criteria and access typical analysis scripts without re-writing or developing having to re-write them or develop new ones. This has enabled groups across WMG to access a large data pool and standardised analyses, which they would not have otherwise had access to. Having a uniform data curation tool has also been a key component of WMG's data strategy, as it has ensured good use of collated data on various types of lithium-ion cells whilst complying with regulatory requirements. There is enormous potential to utilise this data and access the system commercially, and several companies and other research institutions have already expressed interest.

WMG's studies into battery ageing have benefitted the UK automotive industry by providing modelling solutions for them. These solutions de-risk warranty decisions without having to conduct lengthy testing during product development. The data covers a wide variety of test cases – something the automotive industry would have struggled to achieve alone – and results in a huge improvement in the confidence of outcomes. A follow-on project from COBRA, entitled VIPER (Validated and Integrated Platform for battEry Remaining useful life), was secured by WMG and funded through Faraday Sprint to the value of £220k to work closely with Eatron and supported commercialisation of the Remaining Useful Life algorithms.

Several state-of-the-art Machine Learning approaches are now being investigated to improve model generalisability for different cell chemistries and formats. As WMG's data repository expands over time, models will be available for different cell chemistries and use cases, advancing understanding in this area. Over time and as the data repository expands, models will be available for different cell chemistries, use cases, and form factors.

Conclusion

The precision-ageing facility has enabled WMG to enhance its capability and secure additional grants from Innovate UK and the Faraday Institution. These would not have come about had WMG not conducted testing and curated an extensive battery ageing data set useful for both industry and academia. This research project has been instrumental in advancing research and innovation in support of the UK battery industry.

Find out how WMG can help you start your own success story. Contact us at wmgbusiness@warwick.ac.uk